Different readings from barebones and cased Sawbird H1

[Cathal O'Donghaile]

Hi Everyone

I'm hoping someone can help me. I've been testing both a barebones H1 and cased H1 filter from Sawbird with AirSpy and I'm getting very different readouts. Most notably, I'm getting a faint hydrogen reading from the barebones and nothing from the cased one. Attached are screenshots:

So 3 things stand out:

• Hydrogen bump on Barebones
• The difference in profile between the 2 devices
• The cased H1 signal is louder sitting above the 50Db line whereas the barebones signal is slight quieter below the 50Db line
  
  Anyone any ideas what might be wrong? Is it possible the cased one isn't picking up the signal from the probe. Just a note I have a copper probe attached.

Would appreciate the help, thank you

Cathal

[Cathal O'Donghaile]

Attached pics once again. Not sure if they are displaying above.

[Alex P]

I can't comment on the differing plots BUT, in either case your settings need to be changed.

use a 512 FFT 

9 million averages requires ~ 50 Minutes of Time .. try  Intermediate Avg 100 & a total of 300K = 100 seconds

only that tiny shift looks like H data

Disable Offset Tuning

[edhar...@gmail.com]

Obvious questions:

 

How are you powering the cased Sawbird?  (Is the LED lit?)

Are you powering via the USB connector or via the RF line/SDR?  (Does your SDR actually support power?)

(As it’s too easy to do… I’ve done it… :o) Are you sure you’ve got the antenna on the input side and SDR on the output side?

 

Ed

 

From: sara...@googlegroups.com <sara...@googlegroups.com> On Behalf Of Cathal O'Donghaile
Sent: Monday, May 11, 2026 10:17 AM
To: Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Subject: [SARA] Re: Different readings from barebones and cased Sawbird H1

 

Attached pics once again. Not sure if they are displaying above.

On Monday, 11 May 2026 at 15:09:09 UTC+1 Cathal O'Donghaile wrote:

Hi Everyone

 

I'm hoping someone can help me. I've been testing both a barebones H1 and cased H1 filter from Sawbird with AirSpy and I'm getting very different readouts. Most notably, I'm getting a faint hydrogen reading from the barebones and nothing from the cased one. Attached are screenshots:

 

 

So 3 things stand out:

 

• Hydrogen bump on Barebones
• The difference in profile between the 2 devices
• The cased H1 signal is louder sitting above the 50Db line whereas the barebones signal is slight quieter below the 50Db line
  
  Anyone any ideas what might be wrong? Is it possible the cased one isn't picking up the signal from the probe. Just a note I have a copper probe attached.

Would appreciate the help, thank you

 

Cathal

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sara-list/b65f3e71-3d83-48e5-ae8d-71bb76f8ede1n%40googlegroups.com.

[b alex pettit jr]

Use a 50 Ohm Terminator on the LNA input for Background Correction acquisition

It will give you a  flat line across the IFavg Spectrum & data files

[Cathal O'Donghaile]

For Ed:

Yes, the LED is lit up on both the cased and barebones. I'm providing power via the coaxial between the SDR and LNA. It's the NooElec Smartee V3.  I've also made the mistake of wiring the wrong way around, but this time no lol.

Alex:

Thanks for those settings and the suggestion of 50 Ohm terminator. I'll re-run tests. 

[b alex pettit jr]

Cathal,

Your current settings requires 50 mins to capture a Background Correction frame.

THEN, when you begin to capture Real Data, IFaverage tosses the first frame before recording a file...

meaning you may be waiting 1 1/2 hrs or more to capture One File

[b alex pettit jr]

I should have added : it takes at least 50% of the frame time for the spectrum to begin to stabilize & show the final value.

Meaning, you need to wait at least 1/2 hr before really 'seeing' any data with your current setting ....  so go with much

shorter  time frames during eval.

===================================================

[Cathal O'Donghaile]

Thanks, Alex. I've recalibrated the background and set the IF Average parameters per your guidance. The background is now completely flat. 

I'll be taking new readings in the morning. See how it goes.

[b alex pettit jr]

OoooooK ! 

[Stephen Arbogast]

Let's pretend that we live in  a  perfect  universe  where there is no rfi  or noise to  make our measurements   difficult.  This would be boring in my  opinion.  Just  my humble  opinion.   There are many  hardware and  software   choices in  amateur  radio astronomy for hydrogen line  observations.  A few of  mine...

Hardware;

RTlSDR  v3...    very  good

AirSpy   .. very good

ADALM Pluto... over kill  for   Hydrogen line

HackRF... not  a good receiver   

Software;

ezRa....  my  favorite!

SDR#...  with  IFAverage  plugin  works well  on  Windows

VIRGO

DISPIRA

SDRAngel  is  very  cool   but    I  currently  don't  have   post  processing  software  that can handle  it's  output  .csv file.

Recently  I have been  playing with  ideas  in  python  that I have  learned from other   contributors.

Keep  exploring and learning!!!!!

Stephen

[Stephen Arbogast]

Forgot to mention....  my  current  hardware/software  is   Discovery  Dish  and  Discovery  Hydrogen  Line  feed with   ezRA software...   But     I am not  stuck on this  software ....  I am experimenting   with   other  software  such as  SDRAngel...   I recently  did  a  24 hour  drift scan   and got  excellent  averaged  spectrum  using   200,000  samples,  about  2.7  minutes  on my mini pc...  Like   a  dummy  I did not take a  screen shot  during  the  best  galactic  plane  crossing.

I have  been playing  with  collecting   spectral samples and  then  processing them using   a  Jupyter notebook as Eduard  Moll   awakened me to.   This  is  very cool stuff   for   trying    ideas......

So  explore..   try  different  stuff!

Stephen

[Stephen Arbogast]

I have  ordered   the Discovery  Drive  so   I can  easily   point  my  dish  to  ground/cold  sky  for    calibration    hot/cold  but won't ship until   August.

[Cathal O'Donghaile]

So, after extensive tests and recalibrating the background recording here are the results:

Cased_test: Completely flat reading as expected as it wasn't pointed towards the Milky Way

Barebones_Test: I put the LNA inside a hard drive caddy enclosure in an effort to remove RFI. That curved profile is still there but not quite as pronounced as it was.

Barebones_Test_2: I'm getting a pretty decent hydrogen bump on this one while it is pointing towards the Milky Way.

Cased_Test_2: Still completely flat.

I suspect the cased LNA isn't picking up a signal from the prob at all. I'll look at getting onto NooElec about it and see can I run some kind of test on it. Anyone else experience this problem? I've attached a pic too of the LNA with the hard drive caddy enclosure.

Cathal

[b alex pettit jr]

Hello Cathal,

That is still a very low signal level  .

Even a bare Cantenna feed pointed skyward has nearly 2X the levels you are obtaining.

[Cathal O'Donghaile]

This is my horn antenna

[b alex pettit jr]

I processed data from Jan Lustrup using a Horn somewhat similar in size to yours .

They normally perform Extremely Well.

Alex Pettit

[Cathal O'Donghaile]

Thanks, Alex. 

I've much stronger readings from it before. One is attached from Ezra. It got knocked over in a storm and I had to rebuild it. Only now I'm getting some reading from it, but nothing like from before.

Cathal

[b alex pettit jr]

 When you find the problem and get some stronger signals, send me a set of IFaverage data 

& I'll  process it via HL3D > Rinearn and send you the results.

I can't eval Ezra data as I am unsure of the data scaling.

Alex

[Cathal O'Donghaile]

Just an update on this that NooElec are sending me a replacement LNA as it seems to them the one I have doesn't work.

[b alex pettit jr]

There is not much signal strength with the Other either ..

Cathal, 

Microwave radiation has very short wavelengths as compared to AM FM TV radio stations. 

On those longer wavelengths, sometimes a long wire works okay.  

Microwave wavelengths are beginning to have characteristics of Light .. 

Consider how well your "horn' would be useful for such 

Take some time  in the fabrication of your Horn ..

Especially the length of the 1/4wl feed (  near 49mm overall ) and its interfacing to the conductive inner surface of the can.

If your goal was to confirm you

 " Built a radio telescope and received  hydrogen radio frequency emission from the Milky Way "

             The Ezra plot confirms such .. 

To Improve that signal level .. maybe a  rework should be considered .

-Alex

=============================================================

[Cathal O'Donghaile]

Update. Replacement unit works and am getting hydrogen bump. However, it isn't strong. I will work on your suggestions, Alex.

[b alex pettit jr]

Yup ! a good beginning !

Thanks for the update,

Alex

[Cathal O'Donghaile]

[Cathal O'Donghaile]

Stronger bump here after I repositioned slighty

[b alex pettit jr]

Yes 

Use this chart, Stellarium, and something as simple as this and a compass for positioning

Different antenna, but you get the idea

Example if  you lived at 28deg 42mins N Latitude 

always along your Meridian ( N<>S line overhead )

[Andrew Thornett]

I recommend a digital inclinometer for elevation - very cheap and very accurate and apps available free on modern phones to do the function.

Andy

Sent from Outlook for Android

---

From: 'b alex pettit jr' via Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Sent: Tuesday, May 19, 2026 12:16:19 PM
To: sara...@googlegroups.com <sara...@googlegroups.com>
Subject: Re: [SARA] Re: Different readings from barebones and cased Sawbird H1

 

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/sara-list/1045816219.2212947.1779189379790%40mail.yahoo.com.

[Cathal O'Donghaile]

I've been running rtl_power and converted to Ezra thanks to Ted's script. Reading from this morning. Quite a different profile from before. Previously for a probe, I was using a piece from a telescopic aerial, but it was hollow. I replaced it with a piece of solid copper from an old electric wire. 

[b alex pettit jr]

A hollow tube is Fine  : I use such. Maybe it was not properly soldered ?

HLine data changes Smoothly and Slowly

These discontinuities are not correct   

One Sample per Minute is Way Too Fast.  try 4 minutes per sample

===============================================================================
>>> To Evaluate the Performance of a System Use   SDR# > IFavg > HL3D > Rinearn

================================================================================

Alex

[Cathal O'Donghaile]

I will adjust to 4 mins. Thanks, Alex. Yes I was a little worried about the break. Looks like RFI. Here's a graph of the rtl_power output. That spike circled is like 1.5db and coincides with the break. Am I still getting RFI? 

[Cathal O'Donghaile]

I've put the readings through ezSky to map it to Galactic map.

[b alex pettit jr]

Yes, the "RMS" value probably varies similar to the overall signal drift due to environmental changes.

Kinda like these spectra shifts  24 hrs of data

Drift ( background shifts )

The HL3D code corrects for cold sky ( Non HLine ) shifts

& the Amplitude is in validated Engineering Units : dB re Cold Sky

Sources of the changes ? ..  I have a list .. will post later

Alex

=======================

[b alex pettit jr]

[Cathal O'Donghaile]

Thanks, Alex for the tip of taking reading every 4 mins. Much brighter now and not split up like before. However, there is still some RFI in there. I'm studying your list of sources and there are a couple I am exposed to:

• Sky Transparency - heavy cloud here the last few days, but days ahead will be clearer
• Outside temperature variation - warm days and cold nights at the moment. That's about to change to warm days and warm nights
• Possible sunlight refraction off clouds.

I'll know over the coming days. I've to do some solder work to fix the probe too. Ordered up materials to do that.

Thanks for all the help, guys.

Cathal 

[b alex pettit jr]

Hello Cathal,

THAT data is quite good.

I would like to compare EZcon  to   SDR# > IFavg >HL3D

Could you perform 8 to 24 hour acquisitions at the same Declination  using

both EZcon  AND  SDR# > IFavg ( each at 4 mins per sample ) ?  ( obviously on 2 different days  )

Send me the IFavg data files & I'll process them via HL3D and send you the results.

FYI : HL3D now has extremely effective RFI removal tools that Do Not Change the HLine data

AND, since this is  Post-Test processing software,  you can experiment with the settings.

Congrats on the great progress !!

Regards,

Alex

==========================================

[Cathal O'Donghaile]

Yes, I will do that, Alex. The EzCon one will run the rest of the day today. For the IFavg, I'm guessing I need to time how long it takes to do the 901000 readings and then just multiply that out to the period I want to gather, yes?

[b alex pettit jr]

I timed the nooelec SMARTee SDR : it takes 360K per minute .. 4 mins = 1520000  samples   ( 1.52 million ),

but yes, time it to be sure .. not critical .. just around 4 mins.

= Sounds Like a PLAN  Thanks.

========================

[Cathal O'Donghaile]

I left it running through the day. This dark area is rather interesting. 

[b alex pettit jr]

The plot below is more "interesting" than useful as the scaling corresponds to an arbitrary color range,

and I'm not sure  what processing is used for this: I've never seen this detailed.

Finish this & try again with SDR# >  IFavg ..  When processed with HL3D, it will produce plots

Scaled in Engineering Units vs Colors and this can be used for performance evaluation.

HL3D

Y axis H1 signals in Engineering Units and within the spectra you can see RFI, etc.

Alex

=========================================

On Friday, May 22, 2026 at 11:15:31 AM EDT, Cathal O'Donghaile <cathalod...@gmail.com> wrote:

[Cathal O'Donghaile]

I'll be doing the SDR# data collection in the coming days, Alex and I will send you the results. Want to see what happens the readings under clear skies over the weekend with EzCon first.

[b alex pettit jr]

Although I have seen a few short ( 30 sec ) changes when the Sun is  near the beam pattern and small cumulus clouds pass by,

Overall I have Never seen any  performance difference between Clear/Cloudy/PartlyCloudy Days or Nights.

Alex

===================

[tedcl...@gmail.com]

Delighted to see your progress !

With that 12 hour drift-scan, I can start to see the strong signal slowly varying in frequency, the "Galactic hydrogen snake", due to the Earth's rotation and its speed traveling around the Sun.

And near sample 18, I see a faint second power bump at a higher frequency, perhaps the sign of the Outer Galactic arm ?

I think that "dark area" on your plot is due to using the default settings with no collected Reference samples.

That "dark area" would be a calculation "shadow" of using the first sample as the one and only Reference sample.

Does your ezCon047 AntB plot look better ?  (AntB does not use Reference samples)

Or try maybe try re-running the ezCon analysis, using that quiet sample 72 as the only Reference sample,

      py  ..\ezRA\ezCon.py  data\2026-05-22T054902.txt  -ezConRefMode  -72

It is time to collect longer 24-hour drift-scans.

The   ezColIFAvg.py   program can translate "SDR# > IFavg" data format files to ezRA .txt data format files.

Have fun.

---
Ted Cline
TedClineGit at gmail.com

On Friday, May 22, 2026 at 9:15:21 AM UTC-6 Cathal O'Donghaile wrote:

...

[Cathal O'Donghaile]

Thanks, Ted! Very happy with my progress. So, here's the 047 without specifying a sample. It reveals the galactic snake indeed nice and bright.

And I ran your script specifying sample 72 and here's 067:

I've run ezSky too and here's the map:

[Cathal O'Donghaile]

Ted, thought you might like the ezCon file to play with

[Cathal O'Donghaile]

I've done my 24 hour drift scan and also tried something to reduce RFI.

As you can see, I am getting a much clearer plot than before and it also looks like the energy reading at sample 156 is still there and a little more pronounced than before. Ted, would it be the Galactic Outer Arm?

 

[Cathal O'Donghaile]

So, I put the Raspberry Pi and SDR inside an anti-static bag for a motherboard. Previously, the SDR sat outside the box out in the open. Some RFI still remains, but it's a much clearer plot and I'm thrilled to see it.

I've attached the EzCon file here too.

I'll be lining up the SDR## test in the coming days. 

[tedcl...@gmail.com]

Hi Cathal,

I like your latest purple plot above.

I think the ezCon analysis worked well because your data file happens to start with a quiet first sample.

Otherwise, without any Reference Samples, use that   -ezConRefMode   parameter to select a quiet sample.

Without any Reference Samples, maybe use    -ezConAntXInput  4   to use the AntB signal.

Yes, I think the faint top blob, at around your sample 156, comes from the Outer Arm.

The research papers cannot agree on that arm's name (Outer Arm, Norma Arm, Norma-Cygnus Arm),

      https://grokipedia.com/page/norma_arm

      https://en.wikipedia.org/wiki/Norma_Arm

ezGal chose to plot the Sun below the Galactic Center, but this rotated artwork is common on the web,

      https://www.bibliotecapleyades.net/imagenes_universo/cosmos15_16_small.jpg

I think this artwork labels the Galactic Arms more clearly,

      https://www.researchgate.net/profile/Judith-Giannini/publication/369089233/figure/fig1/AS:11431281125473124@1678314583909/Schematic-map-of-the-spiral-arm-structure-of-the-Milky-Way-adapted-from-the-NASA-image.png

ezGal plots the bright Perseus Arm with a fainter Outer Arm, along with some Inner Arms.

And my top link above extends that "Outer Arm" to a "New Outer Arm" found in another direction.

      

Neat !

You have been busy with this for years now.

What did you change recently, to achieve such success ?

---
Ted Cline
TedClineGit at gmail.com

[Cathal O'Donghaile]

Thanks for the feedback, Ted! A few things I changed recently:

• Parts had to be replaced. Last year when I finally got it working, the unit was damaged in a storm one night, including the LNA which I had not previously weatherproofed properly. Also, the boards for the waveguide needed to be replaced. Over the last 2 months I managed to replace everything. However, the new LNA I bought didn't work from the outset. NooElec replaced it and I finally got that bump on the IfAvg. I've also weatherproofed it.
• Made the structure more sturdy. I made the wooden frame to make the whole structure more sturdy and also protect it from the weather.
• Changed the probe. Previously I was using hollow aluminium probes. I'm now using a solid copper probe. 
• Cantenna now attached.  Previously, the waveguide just "sat" on top of the cantenna without it being physically attached to it. This created the readings I got May - August last year.  I was able to attach the cantenna to the wave guide recently which made it more aligned with the wave guide than before.
• Changed the timing on collection of samples. I'm using rtl_power to take the readings. Previously it was set to 60 seconds. At Alex's recommendation, it's now 240 seconds. This created a much cleaner and brighter reading.
• Put SDR and Pi inside anti-static bag. This appeared to reduce RFI though, after you pointed out that I started recording during a quiet period, would like to get the same result starting at around 6am. I'll do this after I do the SDR# test.

Ted, Alex, your help has been invaluable. Thank you so much, guys, couldn't have done it without you. Onwards and upwards.

Cathal

[Cathal O'Donghaile]

Ran the script using samples 18 and 77 as reference. Here's the result

Changed it somewhat (dark purple looks to be gone) but largely same.

[b alex pettit jr]

It is most important to evaluate the spectral data set (plot1) to ascertain RFI, and show the drift error in the set (plot2)

   HL3D automates the identification of the 'lower' cold sky region and normalizes the data set to that value. (plot3)

     The result is a nice, high quality 3D plot (plot4).

===========================================================================

>>  RAW Data << from above scan  flipped to match your plots  = data upon which plot4 above is based

Alex Pettit

=======================================================

[Cathal O'Donghaile]

Thanks, Alex. I'll be running a Hline data collection today for 6 hours between midday and 6pm to get Crab Nebula and upload the results here. Then I'll set it up to run through the night and capture Cygnus A.

Cathal

[b alex pettit jr]

Hello Cathal,

These will be the most interesting plots in characterizing the system :

System Stability  ( combination of hardware, environmental,  RFI )

System Performance ( in calibrated units so that the results can be compared to other systems )

Alex

==================================================

[b alex pettit jr]

Cathal,

At what Declination is your antenna positioned for this ? 

And at your Meridian ?

Alex

[Cathal O'Donghaile]

According to a converter: Declination: -1.38° 

My meridian is 7.1W

I hope that's right

[b alex pettit jr]

Declination: -1.38°  is probably not the best for system evals = its a weak area of the sky ..

Use Stellarium https://stellarium.org/

Take scans at  Dec +40 degrees for max signal strengths.

There are Huge variations of HLine intensity at different Declinations ( and of course RA )

"Meridain" is the imaginary NS line that always passes Overhead

[Cathal O'Donghaile]

Apologies, Alex, I had that wrong. My Dec is 40.

[Cathal O'Donghaile]

Hi Alex. Here's my h-line set of files. They cover about 6 hours when the Crab Nebula passed by. 

[Cathal O'Donghaile]

Ted, what's the usage for your  ezColIFAvg.py script? I've got the data on a USB drive with text files in a h-line folder.

[b alex pettit jr]

Some quick plots ..

Raw Data set : plots of IFavg files with NO processing  ( units of "power"  ..not too meaningful )

This data is RFI Noise Free ... 

Drift  Error vs Hrs time ( seems higher  than normal )

Normalized Data  

3D

Very Useful to eval your current system & to compare to future improvements :)

Alex

=======================

[b alex pettit jr]

another plot

[Cathal O'Donghaile]

Thank you so much, Alex. I love these plots!! Especially the 3D one. Also the news I'm RFI free! A static bag go figure! Might try do some plots myself from the hline files. 

[b alex pettit jr]

Hello Cathal,

Glad you like the plots .. 

The 3D are the "Cake&Ice_Cream"

The "Meat&Potatoes" are the 2D = they show the performance characteristics of the system.  

Yes, experiment w/ HL3D & Rinearn.

There are some info files on HL3D .. it is pretty  simple ... no lines of code to change, just a few variable entries on the GUI

HL3D Info
https://github.com/AP-HLine-3D/HLine3D/blob/main/21cm_Step3A_HL3D_Processing_mx%2Bb_Rev013.pdf

Rinearn Graphics Info

https://github.com/AP-HLine-3D/HLine3D/blob/main/21cm_Step3B_HL3D_Rinearn_Graphics_Rev001.pdf

Once you get a good data set, small changes in the Rinearn Range settings can make a significant change in the presentation.

Attached are  a few of the HL3D processed files for you to plot .

Alex

=============================================

[tedcl...@gmail.com]

Your new "SDR# > IFavg" data format files do not contain pointing info, nor UTC time.

Above, you say your   h-line   directory starting with an   h1line_cathal_0001.txt   file

was a drift-scan from "the Crab Nebula".

I think the Crab Nebula is at about +23 degrees Declination.

Your Earth location is Lat 52.3 Long -7.1, so

    this data's Elevation is 90 - (52.3 - 23) = 60.7

I put your many   h-line.zip   data files in an   ezRABase\CathalODonghaile\data\h-line   directory,

and I ran this (Windows) data format translation command,

      py  ..\ezRA\ezColIFAvg.py  -ezRAObsName  WaterfordIreland  -ezRAObsLat  52.3  -ezRAObsLon  -7.1

               -ezRAObsAmsl  100  -ezColElevation  60.7  data\h-line

to create an   ezRABase\CathalODonghaile\data\h1line_cathal_0001.txt

ezRA   .txt   data file.

But I think your data recorded the PC's Ireland local time, not UTC, so I ran this one-hour time change command,

      py  ..\ezRA\ezFix.py  -ET -01:00:00  data\h1line_cathal_0001.txt

to create a new   ezRABase\CathalODonghaile\data\h1line_cathal_0001b.txt

ezRA   .txt   data file.

If I ran ezFix again it would not step on the previous data.  It would create an   h1line_cathal_0001c.txt   file.

I ran these analysis commands,

      py  ..\ezRA\ezCon.py  data\h1line_cathal_0001b.txt

      py  ..\ezRA\ezSky.py             h1line_cathal_0001b.ezb

But I will say that the ezSky200 plot looks better with a two-hour time change, using   -ET -02:00:00   .

Helpful ?

---
Ted Cline
TedClineGit at gmail.com

On Sunday, May 24, 2026 at 11:29:37 AM UTC-6 Cathal O'Donghaile wrote:

Ted, what's the usage for your  ezColIFAvg.py script? I've got the data on a USB drive with text files in a h-line folder.

...

[Cathal O'Donghaile]

Thanks, Ted, I ran that conversion and it produced interesting graphs, most notably in the AntBTV Doppler. I'll put up 2 sets here - one from IfAvg conversion and another from rtl_power conversion for the same time period on both days.

From h-line:

from rtl_power:

As you can see the 067 from h-line is brighter but also the doppler is quite different and noticeably stronger. I'm wondering is this related to using IfAvg to remove background noise? Do I need to do something similar to the rtl_power output?

Just a note, I didn't compensate on the h-line for UTC times. They are both IST. I wanted the same time ref on them both.

Cathal

[b alex pettit jr]

For Reference,

Neither IFaverage nor HL3D processing clips the  'noise floor'

( this plot shows the full band of frequencies actually acquired by the SDR# settings )

On Monday, May 25, 2026 at 03:59:05 AM EDT, Cathal O'Donghaile <cathalod...@gmail.com> wrote:

Thanks, Ted, I ran that conversion and it produced interesting graphs, most notably in the AntBTV Doppler. I'll put up 2 sets here - one from IfAvg conversion and another from rtl_power conversion for the same time period on both days.

From h-line:

from rtl_power:

[b alex pettit jr]

Same data with HL3D filtering = minimal effect on data 

[Cathal O'Donghaile]

Thanks for that clarification, Alex. I was thinking of the "Acquire background" part of the plugin. What does that algorithm do with the data?

[b alex pettit jr]

the Background Correction removes This, which, if done correctly, to a nearly straight line

( this is the result of the band pass filter ripple and makes Seeing the Data during acquisition quite challenging )

~ RA1200

~ RA2030

Alex

[Cathal O'Donghaile]

Ahha! Okay. I might do an experiment and run rtl_power with the 50 Ohm resistor attached to the LNA input and see what comes out the other end. 

Cathal

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to a topic in the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sara-list/KciwxQ5V4uI/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sara-list+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sara-list/869218220.3398510.1779705231870%40mail.yahoo.com.

[Cathal O'Donghaile]

I asked Google to identify the faint reading from the 23rd:

This was the reply:

What do you all think?